aboutsummaryrefslogtreecommitdiff
path: root/lib
diff options
context:
space:
mode:
authorBen Pfaff <blp@nicira.com>2011-04-01 10:50:52 -0700
committerBen Pfaff <blp@nicira.com>2011-04-29 14:31:50 -0700
commit95a224c5f4ea3558f67c9f9f28e6e8309d8b8c9a (patch)
tree253b5ca8764d0ff7843db62937194bd9b4148fc3 /lib
parentcb447687e9c37bc5ecf4bddc2a2c2e744becc16d (diff)
ovsdb-idl: Fix atomicity of writes that don't change a column's value.
The existing ovsdb_idl_txn_commit() drops any writes that don't change a column's value from what was last reported by the database. But this isn't a valid optimization, because it breaks the atomicity of transactions. Suppose columns A and B initially have values 1 and 2. Client 1 writes value 1 to both columns in one transaction. Client 2 writes value 2 to both columns in another transaction. The only possible valid results for any serial ordering of transactions are 1,1 or 2,2. But if both clients drop writes to columns that they have not modified, then 2,1 also becomes possible (because client 1 just writes to B and client 2 just writes to A). However, for write-only columns we can optimize this out because the IDL can assume it is the only client writing to a column. Found by inspection.
Diffstat (limited to 'lib')
-rw-r--r--lib/ovsdb-idl.c19
1 files changed, 16 insertions, 3 deletions
diff --git a/lib/ovsdb-idl.c b/lib/ovsdb-idl.c
index b264591d..911d7b88 100644
--- a/lib/ovsdb-idl.c
+++ b/lib/ovsdb-idl.c
@@ -1436,9 +1436,7 @@ ovsdb_idl_txn_commit(struct ovsdb_idl_txn *txn)
&class->columns[idx];
if (row->old
- ? !ovsdb_datum_equals(&row->old[idx], &row->new[idx],
- &column->type)
- : !ovsdb_datum_is_default(&row->new[idx],
+ || !ovsdb_datum_is_default(&row->new[idx],
&column->type)) {
json_object_put(row_json, column->name,
substitute_uuids(
@@ -1633,6 +1631,21 @@ ovsdb_idl_txn_write(const struct ovsdb_idl_row *row_,
assert(row->old == NULL ||
row->table->modes[column_idx] & OVSDB_IDL_MONITOR);
+ /* If this is a write-only column and the datum being written is the same
+ * as the one already there, just skip the update entirely. This is worth
+ * optimizing because we have a lot of columns that get periodically
+ * refreshed into the database but don't actually change that often.
+ *
+ * We don't do this for read/write columns because that would break
+ * atomicity of transactions--some other client might have written a
+ * different value in that column since we read it. */
+ if (row->table->modes[column_idx] == OVSDB_IDL_MONITOR
+ && ovsdb_datum_equals(ovsdb_idl_read(row, column),
+ datum, &column->type)) {
+ ovsdb_datum_destroy(datum, &column->type);
+ return;
+ }
+
if (hmap_node_is_null(&row->txn_node)) {
hmap_insert(&row->table->idl->txn->txn_rows, &row->txn_node,
uuid_hash(&row->uuid));